#linuxcnc-devel Logs

Jan 23 2022

06:48 AM smoe: pere: around?
06:55 AM smoe: The consequence from my side is that we need to repeat the .po file generation also for the Spanish.
06:57 AM smoe: Another question that somehow surfaced from the blue - should be not look at the .po files and decide where we need to curate the English a bit more?
06:58 AM smoe: Anything that looks like "cut in half" is where I would want to start. But that could also happen after the initial upload to weblate - it is not too many of these.
07:08 AM pere: I believe the time is better spent on the english original in master to clean up the formatting there, and then in weblate with the production translations of the documentation, instead of more on improving the migration PO files a lot. After all, several thousand strings have been matched already, and looking at them individually should lead to translated documents, not improved migration.
07:08 AM pere: As soon as silopolis[m] say he is done, I am ready to push the new docs build to a PR for review and merge.
07:14 AM smoe: ok - the problem I just ran into is that merging from master keeps throwing conflicts in large chunks of text since the formatting keeps affecting so many lines. So, I keep redoing the reformatting since I cannot tell what smallish change was performed on the master branch. This is a bit difficult. I do not have any splendid idea about how to
07:14 AM smoe: change that, except for limiting the number of changes done to the documentation in the meantime.
07:14 AM smoe: I am admittedly violating that at the moment, but hopefully mostly on files without a translation.
07:29 AM pere: smoe: yes. forget merging from master.
07:31 AM smoe: pere, for the translated files, master is a non-issues, but for the English? I need to get my changes back somehow.
07:31 AM pere: smoe: I suspect you will have to cherry-pick them into a new master based branch.
07:32 AM pere: or perhaps do it manually instead of git cherry-pick
07:32 AM smoe: That is what I have done. Generate a .diff of all English .adoc and then apply this on a new branch off master.
07:33 AM smoe: Yes, I spent a weekend on this, and now another one.
07:34 AM pere: better get those improvments quickly into master before it diverges even further.
07:47 AM smoe: Yes, this needs to be reviewed. It is here https://github.com/LinuxCNC/linuxcnc/pull/1516
07:48 AM smoe: I mean, what we are doing for weblate is basically a mostly formal (typos) review of what is currently in the archive. That should be mostly straight-forward.
07:48 AM smoe: Will in parts something be messed up? Yes.
07:49 AM smoe: Will it be dramatic? No
07:51 AM smoe: We yet do not have a medium where to talk about the various parts of the documentation, and I do not know how to get there.
08:33 AM pere: me neither.
02:06 PM smoe: I think I completed my efforts to review the .adoc syntax now.Just another stale link to fix.
02:09 PM smoe: I also started reading a bit more in the PDFs as provided on the web page. This sounds a bit like bike-shedding but regular asciidoc-tables are a prime candidate to be improved to become more appealing. I read through https://docs.asciidoctor.org/asciidoc/latest/tables/build-a-basic-table/ and got plenty of ideas. Most notably I want to eliminate
02:09 PM smoe: most grid-lines, i.e. for all the sufficiently simple tables that is.
02:10 PM smoe: https://www.scienceeditingexperts.com/blog/creating-tables-in-scientific-papers-basic-formatting-and-titles give a short summary of what I am after.
02:10 PM smoe: Now, with whom can I talk this through? Or should I just provide a PR?
04:31 PM Tom_itx is now known as Tom_L